Abnormality detection and intelligent severity assessment of human chest computed tomography scans using deep learning: a case study on SARS-COV-2 assessment

Background Over half a million individuals are diagnosed with head and neck cancer each year globally. Radiotherapy is an important curative treatment for this disease, but it requires manual time to delineate radiosensitive organs at risk. This planning process can delay treatment while also introducing interoperator variability, resulting in downstream radiation dose differences. Although auto-segmentation algorithms offer a potentially time-saving solution, the challenges in defining, quantifying, and achieving expert performance remain. Objective Adopting a deep learning approach, we aim to demonstrate a 3D U-Net architecture that achieves expert-level performance in delineating 21 distinct head and neck organs at risk commonly segmented in clinical practice. Methods The model was trained on a data set of 663 deidentified computed tomography scans acquired in routine clinical practice and with both segmentations taken from clinical practice and segmentations created by experienced radiographers as part of this research, all in accordance with consensus organ at risk definitions. Results We demonstrated the model’s clinical applicability by assessing its performance on a test set of 21 computed tomography scans from clinical practice, each with 21 organs at risk segmented by 2 independent experts. We also introduced surface Dice similarity coefficient, a new metric for the comparison of organ delineation, to quantify the deviation between organ at risk surface contours rather than volumes, better reflecting the clinical task of correcting errors in automated organ segmentations. The model’s generalizability was then demonstrated on 2 distinct open-source data sets, reflecting different centers and countries to model training. Conclusions Deep learning is an effective and clinically applicable technique for the segmentation of the head and neck anatomy for radiotherapy. With appropriate validation studies and regulatory approvals, this system could improve the efficiency, consistency, and safety of radiotherapy pathways.

Download Full-text

End-to-end lung cancer screening with three-dimensional deep learning on low-dose chest computed tomography

Nature Medicine ◽

10.1038/s41591-019-0447-x ◽

2019 ◽

Vol 25 (6) ◽

pp. 954-961 ◽

Cited By ~ 256

Author(s):

Diego Ardila ◽

Atilla P. Kiraly ◽

Sujeeth Bharadwaj ◽

Bokyung Choi ◽

Joshua J. Reicher ◽

...

Keyword(s):

Computed Tomography ◽

Lung Cancer ◽

Deep Learning ◽

Cancer Screening ◽

Low Dose ◽

Three Dimensional ◽

Lung Cancer Screening ◽

Chest Computed Tomography ◽

End To End

Download Full-text

Indeterminate Pulmonary Nodules at Diagnosis in Rhabdomyosarcoma: Are They Clinically Significant? A Report From the European Paediatric Soft Tissue Sarcoma Study Group

Journal of Clinical Oncology ◽

10.1200/jco.18.01535 ◽

2019 ◽

Vol 37 (9) ◽

pp. 723-730 ◽

Cited By ~ 5

Author(s):

Bas Vaarwerk ◽

Gianni Bisogno ◽

Kieran McHugh ◽

Hervé J. Brisse ◽

Carlo Morosi ◽

...

Keyword(s):

Computed Tomography ◽

Overall Survival ◽

Soft Tissue ◽

Soft Tissue Sarcoma ◽

Pulmonary Nodules ◽

Study Group ◽

Event Free Survival ◽

Free Survival ◽

Chest Computed Tomography ◽

Computed Tomography Scans

Purpose To evaluate the clinical significance of indeterminate pulmonary nodules at diagnosis (defined as ≤ 4 pulmonary nodules < 5 mm or 1 nodule measuring ≥ 5 and < 10 mm) in patients with pediatric rhabdomyosarcoma (RMS). Patients and Methods We selected patients with supposed nonmetastatic RMS treated in large pediatric oncology centers in the United Kingdom, France, Italy, and the Netherlands, who were enrolled in the European Soft Tissue Sarcoma Study Group (E pSSG) RMS 2005 study. Patients included in the current study received a diagnosis between September 2005 and December 2013, and had chest computed tomography scans available for review that were done at time of diagnosis. Local radiologists were asked to review the chest computed tomography scans for the presence of pulmonary nodules and to record their findings on a standardized case report form. In the E pSSG RMS 2005 Study, patients with indeterminate pulmonary nodules were treated identically to patients without pulmonary nodules, enabling us to compare event-free survival and overall survival between groups by log-rank test. Results In total, 316 patients were included; 67 patients (21.2%) had indeterminate pulmonary nodules on imaging and 249 patients (78.8%) had no pulmonary nodules evident at diagnosis. Median follow-up for survivors (n = 258) was 75.1 months; respective 5-year event-free survival and overall survival rates (95% CI) were 77.0% (64.8% to 85.5%) and 82.0% (69.7% to 89.6%) for patients with indeterminate nodules and 73.2% (67.1% to 78.3%) and 80.8% (75.1% to 85.3%) for patients without nodules at diagnosis ( P = .68 and .76, respectively). Conclusion Our study demonstrated that indeterminate pulmonary nodules at diagnosis do not affect outcome in patients with otherwise localized RMS. There is no need to biopsy or upstage patients with RMS who have indeterminate pulmonary nodules at diagnosis.

Download Full-text

Multi-Model Ensemble Deep Learning Method to Diagnose COVID-19 Using Chest Computed Tomography Images

Journal of Shanghai Jiaotong University (Science) ◽

10.1007/s12204-021-2392-3 ◽

2021 ◽

Author(s):

Zhiming Wang ◽

Jingjing Dong ◽

Junpeng Zhang

Keyword(s):

Computed Tomography ◽

Deep Learning ◽

Learning Method ◽

Model Ensemble ◽

Chest Computed Tomography ◽

Computed Tomography Images

Download Full-text

Incidence and Implication of Coronary Artery Calcium on Non-gated Chest Computed Tomography Scans: A Large Observational Cohort

Cureus ◽

10.7759/cureus.6218 ◽

2019 ◽

Author(s):

Charles Haller ◽

Anthony Vandehei ◽

Raymond Fisher ◽

Joshua Boster ◽

Brian Shipley ◽

...

Keyword(s):

Computed Tomography ◽

Coronary Artery ◽

Coronary Artery Calcium ◽

Chest Computed Tomography ◽

Computed Tomography Scans ◽

Observational Cohort

Download Full-text

Hybrid Convolutional Neural Network-Based Diagnosis System for Intracranial Hemorrhage

BRAIN. BROAD RESEARCH IN ARTIFICIAL INTELLIGENCE AND NEUROSCIENCE ◽

10.18662/brain/12.4/236 ◽

2021 ◽

Vol 12 (4) ◽

pp. 01-27

Author(s):

Sezin Barin ◽

Murat Saribaş ◽

Beyza Gülizar Çiltaş ◽

Gür Emre Güraksin ◽

Utku Köse

Keyword(s):

Computed Tomography ◽

Deep Learning ◽

Early Diagnosis ◽

Intracranial Hemorrhage ◽

Hybrid Method ◽

Computer Aided Design ◽

Cranial Computed Tomography ◽

Critical Problem ◽

Computed Tomography Scans ◽

Aided Design

Early diagnosis of intracranial hemorrhage significantly reduces mortality. Hemorrhage is diagnosed by using various imaging methods and the most time-efficient one among them is computed tomography (CT). However, it is clear that accurate CT scans requires time, diligence, and experience. Computer-aided design methods are vital for the treatment because they facilitate early diagnosis of intracranial hemorrhage. At this point, deep learning can provide effective outcomes through an automated diagnosis way. However, as different from the known solutions, diagnosis of five different hemorrhage subtypes is a critical problem to be solved.This study focused on deep learning methods and employed cranial computed tomography scans in order to detect intracranial hemorrhage. The diagnosis approach in the study aimed to detect five subtypes of hemorrhage. In detail, EfficientNet-B3 and ResNet-Inception-V2 architectures were used for diagnosis purposes. Eventually, the study also proposed a two-architecture hybrid method for the diagnosis purpose. The obtained findings by the hybrid method were evaluated in terms of a comparative perspective.Results showed that the newly designed hybrid method was quite effective in terms of increasing classification rates of detecting intracranial hemorrhage according to the subtypes. Briefly, an accuracy of 98.5%, which is higher than those of the EfficientNet-B3 and the Inception-ResNet-V2, were obtained thanks to the developed hybrid method.

Download Full-text

Role of novel deep-learning-based CT used in management and discharge of COVID-19 patients at a “square cabin” hospital in China

10.21203/rs.3.rs-28201/v1 ◽

2020 ◽

Author(s):

Qingcheng Meng ◽

Wentao Liu ◽

Xinmin Dou ◽

Jiaqi Zhang ◽

Anlan Sun ◽

...

Keyword(s):

Computed Tomography ◽

Deep Learning ◽

Nucleic Acid ◽

Volume Ratio ◽

Respiratory Pathogen ◽

Normal Lung ◽

The Novel ◽

Chest Computed Tomography ◽

Nucleic Acid Test ◽

Acid Test

Abstract Background: The chest computed tomography (CT) had been used to define the diagnostic and discharge criteria for COVID-19. However, it is difficult to determine the suitability for discharge of a patient with COVID-19 based on CT features in a clinical setting. Deep learning (DL) technology has demonstrated great success in the medical imaging.Purpose: This study applied the novel deep learning (DL) on chest computed tomography (CT) of COVID-19 patients with consecutive negative respiratory pathogen nucleic acid test results at a “square cabin” hospital in Wuhan, China, with the intent to standardize criteria for discharge.Methods: The study included 270 patients (102men, 168 women; mean age, 51.9 ± 15.6[18–65] years) who had two consecutive negative respiratory pathogen tests (sampling interval: ≥1 day) and underwent low-dose CT 1 day after the first negative test, with strict adherence to epidemic prevention standards. The chest CT of COVID-19 patients with negative nucleic acid tests were evalued by DL, and the standard for discharge was a total volume ratio of lesions to lung of less than 50% determined by DL.Results: The average intersection over union is 0.7894. Fifty-seven (21.1%) and 213 (78.9%) patients exhibited normal lung findings and pneumonia, respectively. 54.0% (115/213) involved mild interstitial fibrosis. 18.8% (40/213) had total volume ratio of lesions to lung of more than and equal to 50% according to our severity scale and were monitored continuously in hospital, and three cases of which had a positive follow-up nucleic acid test during hospital observation. None of the 230 discharged cases later tested positive or exhibited pneumonia progression. Conclusions: The novel DL enables the accurate management of COVID-19 patients and can help avoid cluster transmission or exacerbation due to patients with false negitive acid test.

Download Full-text